Gut: first published as 10.1136/gut.26.12.1279 on 1 December 1985. Downloaded from

Gut, 1985, 26, 1279-1283

Leading article Inhibition of acid and gastric

Control of gastric acid secretion has long been one of the principal aims of medical and surgical approach to the management of . Prolonged maintenance treatment with more and more powerful drugs now appears to be on the horizon. Extensive testing of loxtidine, a very potent histamine H2 antagonist which differs from cimetidine and in that it offers non- competitive and insurmountable H2 antagonism' is reported by Dr Poynter and his colleagues in this issue.2 These data and other studies, raise a number of interesting questions about the biology and pathophysiology of the gastric endocrine cells. Poynter et al, using stringent necropsy protocols have found a striking incidence of neuroendocrine gastric tumours of a non-functional variety when loxtidine was administered in high doses to male and female Wistar rats for the whole of their natural life span. The first tumour appeared 712 days after starting the drug and female rats appear to be more susceptible than males. The progressive histological changes ranging from intramucosal proliferation of neuroen- docrine cells to the formation of small tumours and subsequent invasion through the muscularis mucosa, were taken to indicate early . Full malignant potential was evinced in one rat only by replacement of a http://gut.bmj.com/ by malignant cells. These findings are interesting, as spontaneous tumours of the glandular in the rat are very rare and spontaneous carcinoids are not recorded. The authors' hypothesis appears straightforward enough: profound inhibition of gastric acid secretion, which presumably occurred at the doses of loxtidine used, abolished luminal acid inhibition of antral G cells and lead to secondary hy3ergastri- on September 29, 2021 by guest. Protected copyright. naemia which in turn could stimulate enterochromaffin-like cells, produc- ing and eventual neoplasia. There was no evidence of a dose response relationship, however, indicating perhaps that all doses of loxtidine used were well into the supramaximal range for complete acid inhibition. The authors do not report serum or plasma gastrin concentra- tions in this study. Support for their hypothesis comes from the well documented changes seen in idiopathic fundic mucosal atrophy in man where diffuse endocrine cell hyperplasia mainly of enterochromaffin-like cells has been reported.7 Similarly hypergastrinaemia in association with fundic atrophic has also been associated with nodules of endocrine cells termed microcarci- noids.810 Nevertheless, in general, gastric carcinoids are rare, constituting only 3% of all gastrointestinal carcinoids"1-13 and approximately 03% of all gastric tumours.12 A recent review of 123 patients with pernicious anaemia, screened by , disclosed four patients with a solitary, and one patient with multiple gastric carcinoid tumours.14 In these cases serum gastrin concentrations ranged from 1060 to 7500 pmol/l. The tumour 1279 Gut: first published as 10.1136/gut.26.12.1279 on 1 December 1985. Downloaded from

1280 Elder cells stained positively in immunohistochemical analysis using antibodies for neurone specific enolase (a reliable marker for tissue of neuroendoc- rine origin), and this is similar to the observations of Dr Poynter and his colleagues in the rats given loxtidine. Fundic mucosal endocrine cell hyperplasia and hypergastrinaemia were also noted by Carney et al,'5 but whether the proposed series of events leading to endocrine cell hyperplasia and neoplasia is directly associated with hypergastrinaemia is not proven. It is interesting that gastric carcinoid tumours have also been detected in the Zollinger-Ellison syndrome15 16 where serum gastrin concentrations are sometimes higher than in pernicious anaemia. As gastrin is known to stimulate enterochromaffin-like cells17 it is probably that the common factor in the development of gastric carcinoids is hypergastrinaemia and not profound inhibition of gastric acid secretion, as was present in the toxicological study in rats reported by Poynter et al. In some clinical cases of gastric carcinoid, gastric has been found concomit- antly.'1 13 15-21 No suggestion of gastric adenocarcinoma was present in the loxtidine study and this was in contrast to the response in the rat to the histamine H2 antagonists tiotidine22 and SKF93479,23 which after 12 months of high dosage produced of the stomach on the one hand and hyperplasia and hyperkeratosis of the forestomach on the other. Extensive toxicological studies with conventional H2 blockers in the rat did not produce any gastric tumours. Loxtidine itself when tested by the WHO nitrosation procedure yielded no mutagenic products. Moreover, the development of gastric carcinoid tumours is entirely different from that seen after treatment with a nitrosated derivative of an H2-receptor antagonist, which produced adenocarcinoma in an experimentally wounded rat's stomach.24 It is interesting that in experiments with very high doses of ranitidine http://gut.bmj.com/ (2000 mg/kg/day) lasting from 875 to 903 days with plasma gastrin concentrations as high as 8900 mglml Poynter et al, did not find any evidence of enterochromaffin-like cell hyperplasia or carcinoid tumours.25 Was this because ranitidine is a competitive H2 blocker, while loxtidine has a non-competitive action? This observation suggests that factors other than hypergastrinaemia may be involved. In a recent 24 month carcinogenicity on September 29, 2021 by guest. Protected copyright. study using ,26 a proton pump blocker, six of 60 male and 24 of 60 female rats receiving omeprazole 400,umol/kg/day developed gastric carcinoids. The strikingly higher incidence in the female than in the male rats was particularly marked at the lower doses of the drug used in the study (40 and 125 umol/kg/day). A study in mice with omeprazole did not yield focal argyrophilic cell hyperplasia, or carcinoid tumours. These results with a different compound tested in the same species are very similar to the loxtidine study reported in this issue of Gut. The evidence to date suggests that the effects on enterochromaffin-like cells in such models are not dependent on the molecular structure of the drug producing profound inhibition of gastric acid, but are probably associated with the severity and duration of hypergastrinaemia. As it is well known that there is no close correlation between luminal gastric hydrogen ion activity and serum gastrin, it is likely that in man serum gastrin concentration is influenced by other factors, such as vagal stimulation, hypercalcaemia and the presence of antral gastritis as well as by intragastric pH.27-29 An interesting and as yet unexplained finding in the loxtidine and omeprazole Gut: first published as 10.1136/gut.26.12.1279 on 1 December 1985. Downloaded from

Inhibition of acid and gastric carcinoids 1281 experiments and on long term treatment in the rat with a new H2 receptor blocker famotidine,30 was the occurrence of eosinophilic granules in cells adjacent to the chief cells in the basal parts of the gastric glands. A preliminary report has highlighted the importance of the gastric antrum in rats treated with omeprazole (10-400 ,umol/kg orally), or ranitidine (175+175; ,mol/kg orally), compared with antrectomised rats given omeprazole (400 ,umol/kg orally).31 Gastrin concentrations increased in rats with intact stomachs treated with the drugs but they were low in antrectomised controls: detailed studies showed a direct significant correlation between concentrations of plasma gastrin and enterochromaf- fin-like cell density, and with tissue levels of histidine decarboxylase and histamine. These results support the hypothesis that prolonged inhibition of gastric acid secretion and the resulting hypergastrinaemia causes proliferation of the enterochromaffin-like cells.3 There was no evidence of any direct effect of either omeprazole, or ranitidine on enterochromaffin- like cell density and the same is probably true of the studies with loxtidine. All of these studies were done in the rat with very high doses of the drugs. Who needs such profound and prolonged acid inhibitions? From many points of view it is not desirable in man. This is because of the putative dangers of gastric colonisation and formation,32-36 in addition to the risk of gastric carcinoid tumours. Extrapolation of data from rat to the human is difficult as in some ways the rat is less sensitive to the action of some gastrointestinal hormones, for example gastrin, than is man. The production of epithelial and endocrine , however, in association with profound inhibition of gastric acid secretion experimentally recalls Emerson's words: 'Nothing in nature is given, everything is sold'. As in other fields, there is a strong suggestion that nature has done the experiment herself many times in the past,11 15 and it may be that the http://gut.bmj.com/ luminal pH of the stomach has little if anything to do with gastric carcinoids, providing that the hypergastrinaemia reaches as yet an undefined threshold. The ability of antrectomy to prevent these changes in the body and fundus of the stomach is also of fundamental importance in understanding the mechanism of pharmacologically induced gastric carci-

noids, but the question remains why so relatively few clinical cases have on September 29, 2021 by guest. Protected copyright. occurred in those with profound and prolonged hypergastrinaemia, as in pernicious anaemia and in the Zollinger-Ellison . These observa- tions highlight the dangers of extremes and emphasise the need to avoid prolonged excessive suppression, or excessive stimulation of the gastric cells. Are there other modulating factors? In a few patients with gastric carcinoids assays of some 20 peptides including somatostatin have been normal and 24 hour urinary 5-HIAA secretion was also normal in the five patients tested.14 It may be that with treatment for longer and longer periods with increasingly powerful H2 blockers or with omeprazole, which has recently been shown to be very effective for the Zollinger-Ellison syndrome,3 the need for endoscopic surveillance, multiple and argyrophil staining of specimens from these special cases will become much more important. J B ELDER Department of Postgraduate Medicine, University of Keele, Stoke-on-Trent, Staffordshire Gut: first published as 10.1136/gut.26.12.1279 on 1 December 1985. Downloaded from

1282 Elder

References 1 Brittain RI, Jack D, Reeves, JJ, Stables R. Pharmacological basis for the inductionof gastric carcinoid tumours in the rat: loxtidine an unsurmountable histamine H2 receptor blocking drug. Br J Pharmacol 1985; 85: 843-7. 2 Poynter D, Pick CR, Harcourt RA et al. Association of long lasting unsurmountable histamine H2 blockade and gastric carcinoid tumours in the rat. Gut 1985; 26: 1284-95. 3 Hakansson R, Elkeland M, Sundler F. Activation and proliferation of gastric endocrine cells. In: Faulkner S, Hakansson R, Sundler F, eds. Evaluation and tumour of the neuroendocrine system. Amsterdam: Elsevier Science Publications BV, 1984: 371-98. 4 Polak JM, Hoffbrand AV, Reed PI, Bloom S, Pearse AGE. Qualitative and quantitative studies of antral and fundic G cells in pernicious anaemia. Scand J Gastroenterol 1973; 8: 361-7. 5 Rubin W. A fine structural characterisation of the proliferated endocrine cells in atrophic . Ann J Pathol 1973; 70: 109-18. 6 Solcia E, Capella C, Buffa R, Usellini L, Frigerio B, Fontana P. Endocrine cells of the and related tumours. Pathobiol Ann 1979; 9: 163-203. 7 Solcia E, Capella C, Vesallo G, Buffa R. Endocrine cells of the gastric mucosa. Int Rev Cytol 1975; 42: 223-86. 8 Bordi C, Costa A, Missale G. ECL Cell proliferation and gastrin levels. 1975; 68: 205-6. 9 Razenhofer M. Uber enterale Hyperlasien und Geschwulste der Disseminerten Endo- krinen (Parakrinen) Hellen zellen Feyrtes unter Beruchsichtigung Amphikriner Zell- wucherungen. Verh Dtsch Ges Pathol 1977; 61: 7-24. 10 Solcia E, Frigerio B, Capella C. Gastrin cells and related endocrine cells modulating gastric secretion: an ultrastructural study of normal and pathological mucosa. In: Refeld JF, Amdrup E, eds. Gastrin and the vagus. London: Academic Press: 1979: 31-9. 11 Hajdu SI, Winowar SJ, Myers WPL. Carcinoid tumours. Ann J Clin Pathol 1973; 61: 521-8. 12 McDonald RA. A study of 356 carcinoids of the gastrointestinal tract. Ann J Med 1956; 21: 867-78. 13 Godwin JD. Carcinoid tumours: an analysis of 2,837 cases. 1975; 36: 560-9. 14 Borch K, Renvall H, Liedberg G. Gastric endocrine cell hyperplasia and carcinoid tumours in pernicious anaemia. Gastroenterology 1985; 88: 638-48. http://gut.bmj.com/ 15 Carney JA, Go VLW, Fairbanks VF, Moore SB, Alport EC, Nora FE. The syndrome of gastric argyrophil carcinoid tumours and non-antral gastric atrophy. Ann Intern Med 1983; 99: 761-6. 16 Bordi C, Ravazzola M, DeVita 0. Pathology of endocrine cells in gastric mucosa. Ann Pathol 1983; 3: 19-28. 17 Kahlson G, Rosengren E, Svahn D, Thunberg R. Mobilisation and formation of histamine in the gastric mucosa as related to acid secretion. J Physiol 1964; 174: 400-16. 18 Lattes R, Grossi C. Carcinoid tumours of the stomach. Cancer 1956; 9: 698-711. on September 29, 2021 by guest. Protected copyright. 19 Bansal M, Mamuro K, Gordon RE. Carcinosarcoma and separate carcinoid tumour of the stomach. Cancer 1982; 50: 1876-81. 20 Soga J, Tazawa K, Aizawa 0, Wada K, Terukaza T. Argentaffin cell adenocarcinoma of the stomach. An atypical carcinoid. Cancer 1971; 28: 999-1003. 21 Azzopardi JG, Pollok DJ. Argentaffin and argyrophil cells in gastric . J Pathol Bact 1963; 86: 443-51. 22 Streett CS, Cimprich RE, Robertson JL. Pathologic findings in the stomachs of rats treated with H2-receptor antagonist tiotidine. Scand J Gastroenterol 1984; 19: suppl 101: 109-17. 23 Betton GR, Salmon GK. Pathology of the forestomach in rats treated for 1 year with a new histmaine H2-receptor antagonist SK & F 93479 trihydrochloride. Scand J Gastroenterol 1984; 19: suppl 107: 103-8. 24 Elder JB, Wells S, Koffman CG, Ganguli PC, Williams G. Possible role ofcimetidine and its nitrosation product in human . New York: Cold Spring Harbor Laboratory, 1982: 335-49. 25 Poynter D, Pick CR, Harcourt RA et al. Evaluation of ranitidine safety. In: Misiewicz JJ, Wormsley KG, eds. The clinical use of ranitidine. Oxford: Medicine Publishing Foundation, 1982: 49-57. 26 Ekman L, Hansson E, Havu N, Carlsson E, Lundberg C. Toxicological studies on omeprazole. Scand J Gastroenterol 1985; 20: suppl 108: 53-69. Gut: first published as 10.1136/gut.26.12.1279 on 1 December 1985. Downloaded from

Inhibition of acid and gastric carcinoids 1283

27 Elder JB, Ganguli PC. Gastric function. In: Brown SS, Mitchell FL, Young DS, eds. Chemical diagnosis of disease. Oxford: Elsevier, 1979: 663-88. 28 Ganguli PC, Cullen DR, Irvine WJ. Radio-immunoassay of plasma gastrin in pernicious anaemia, without pernicious anaemia, hypochlorhydria and in controls. Lancet 1971; 1: 155-8. 29 Baron JH. Physiology and abnormal secretion and motility: gastric secretion. In: Bouchier IA, Allan RN, Hodgson HJF, Keighley MRB, eds. Textbook of gastrotneterology. London: Bailliere Tindall, 1984: 94-104. 30 Burek JD. Famotidine. A potent histamine H2 receptor antagonist - summary of preclinical safety assessment. Symposium: Famotidine: a further development in the modern treatment of peptic ulcer disease. Lisbon, Portugal, September 19th, 1984. 31 Hakansson R, Sundler F, Carlsson E, Mattson H, Larsson H. Proliferation of enterochromaffin-like (ECL) cells in the rat stomach following omeprazole treatment. Hepatogastroenterol 1985; 32: 48-9. 32 Reed PI, Smith PLR, Haines K, House FR, Walters CL. Effect of cimetidine on gastric juice N-nitrosamine concentration. Lancet 1981; 2: 553-6. 33 Elder JB, Ganguli PC, Gillespie IE. Cimetidine and gastric cancer. Lancet 1979; 1: 1005-6. 34 Stockbrugger RW, Cotton PB, Eugenides N, Bartholomew BA, Hill MJ, Walters CL. Intragastric , and bacterial overgrowth during cimetidine treatment. Gut 1982; 23: 1048-54. 35 Tannenbaum SR. N-nitroso compounds: a perspective on human exposure. Lancet 1983; 1: 629-32. 36 Elder JB, Burdett K, Smith PR, Reed PI, Walters CL. Influence of H2 blocking drugs on endogenous nitrosation in humans as monitored by N-nitrosoproline secretion in the urine. In: N-nitroso compound: report on occurrence and biological effects of N- nitrosocompounds. Lyon: IARC, 1984: 969-74. 37 McArthur KE, Collen MJ, Maton PN et al. Omeprazole: effective, convenient therapy for Zollinger-Ellison Syndrome. Gastroenterology 1985; 88: 939-44. http://gut.bmj.com/ on September 29, 2021 by guest. Protected copyright.